Spatial and Temporal Evolution of the Infiltration Characteristics of a Loess Landslide

Dongdong Yang, Haijun Qiu, Yanqian Pei, Sheng Hu, Shuyue Ma, Zijing Liu, Yan Zhang, Mingming CaoISPRS Int. J. Geo Inf 2020 · DOI

TL;DR

This study investigates how water infiltration changes over time and space in loess landslide bodies, revealing that newly formed landslides have significantly higher infiltration rates than older ones or undisturbed loess. The formation of physical crusts on new landslides dramatically reduces infiltration over several months, suggesting a long recovery period for these areas to reach pre-landslide infiltration levels.

What Problem Does This Paper Address?

Landslides, particularly in loess regions, are often triggered or exacerbated by water infiltration. While the general role of infiltration in slope stability is understood, the specific dynamics of how infiltration characteristics evolve after a landslide, both spatially across the landslide body and temporally over time, remain unclear. This knowledge gap is critical for understanding landslide recurrence, predicting future failures, and assessing the suitability of landslide-prone areas for land use. The paper aims to fill this gap by examining the infiltration rates in undisturbed loess, previously failed landslide bodies, and recently formed landslide bodies.

Methods

The researchers employed a multi-pronged approach:

  • Field Investigations: They conducted site visits to characterize the physical state of the landslide bodies, including the presence of vegetation, biological crusts, and physical crusts.
  • In Situ Infiltration Rate Measurement: A single-cylinder permeameter was used to measure both the initial infiltration rate (IIR) and the quasi-steady infiltration rate (QIR) of the soil. This involved carefully inserting the cylinder into the soil and measuring the volume of water that infiltrated over time.
  • Physical Parameter Measurement: Soil samples were collected near infiltration sites to determine antecedent moisture content, bulk density, and porosity.
  • Temporal Monitoring: For the new landslide body (NLB), infiltration experiments were repeated at different times (October 2017, March 2018, and July 2018) to track temporal changes. This included comparing infiltration rates in areas with and without physical crusts.
  • Statistical Analysis: One-way ANOVA and independent t-tests were used to analyze the differences in infiltration rates, bulk density, and porosity between different sites and over time.

Datasets

The study utilized data from the Baqiao meteorological station for daily and accumulated precipitation records. The primary "dataset" for the infiltration study was generated through in-situ field measurements conducted at various experimental sites on undisturbed loess (UDL), a previous landslide body (PLB), and a new landslide body (NLB) in the Jiwachang Brickyard Landslide area.

Key Insights and Ideas

  • Landslides are Infiltration Hotspots: Newly formed landslide bodies exhibit dramatically higher infiltration rates compared to both older landslide bodies and undisturbed loess. This is attributed to the highly disturbed and fractured nature of the soil.
  • The "Crust Effect": The rapid decrease in infiltration rates in new landslides over a few months is primarily due to the formation of physical crusts on the soil surface. These crusts act as a barrier to water entry.
  • Long Recovery Timescale: It appears that it takes a considerable amount of time (potentially over 10 years) for a disturbed loess landslide body to recover its infiltration characteristics to levels comparable to undisturbed loess.
  • Spatial Heterogeneity within Landslides: Even within a single landslide body, infiltration rates can vary spatially. For instance, the previous landslide body showed higher infiltration in the upper part, intermediate in the middle, and lowest in the lower part.
  • Physical Properties Matter: Lower bulk density and higher porosity in disturbed soils (PLB and NLB) correlate with higher infiltration rates, as expected.

Key Results

  • The average initial infiltration rate (IIR) in the new landslide body (NLB) was approximately 200 times higher than in undisturbed loess (UDL) and 60 times higher than in the previous landslide body (PLB).
  • The IIR in the PLB was about 4.4 times higher than in the UDL.
  • Following landslide formation, the IIR in the NLB sharply decreased by over 98% within 4-5 months due to physical crust formation.
  • In undisturbed loess, infiltration rates were higher in loess layers (L1, L2) than in paleosol layers (S1).
  • In the previous landslide body, infiltration rates were highest in the upper part, intermediate in the middle, and lowest in the lower part.
  • The NLB showed higher infiltration in the lower part and lower infiltration in the upper part.

Limitations and Future Work

  • Crust Formation Complexity: While the study highlights the role of physical crusts, the exact formation mechanisms and duration of these crusts, especially under varying environmental conditions, require further investigation.
  • Long-Term Temporal Evolution: The study provides data for up to approximately 1.5 years after the new landslide. Understanding the infiltration evolution over even longer timescales (years to decades) is crucial for assessing full recovery.
  • Influence of Biological Crusts: The paper notes that biological crusts were observed on the PLB and may contribute to higher infiltration rates compared to UDL, but their specific role and formation timeline on disturbed soils need more detailed study.
  • Generalizability: While the study focuses on a specific loess landslide, further research across different loess regions and landslide types would enhance the generalizability of the findings.
  • Future Work Suggestions: The authors suggest that understanding the formation and duration of physical crusts is essential. They also imply that the long-term recovery of infiltration rates in landslide bodies is a key area for future research, with implications for land use planning.

Summary updated

BibTeX
@article{yang2020spatial,
  author = {Dongdong Yang and Haijun Qiu and Yanqian Pei and Sheng Hu and Shuyue Ma and Zijing Liu and Yan Zhang and Mingming Cao},
  title = {{Spatial and Temporal Evolution of the Infiltration Characteristics of a Loess Landslide}},
  journal = {ISPRS Int. J. Geo Inf.},
  year = {2020},
  doi = {10.3390/ijgi9010026},
}
Figure